List your company

Language Understanding on MMLU (pass@1 and pass@128)

74Pass@1Base
68.870.1571.572.85Apr 13, 2026

Evaluation results

MethodLinks
BaseBackbone=Qwen2.5-7B, Generation budget=4096 tokens2026.047491
GRPO (Countdown)Backbone=Qwen2.5-7B, Generation budget=4096 tokens, Optimization=GRPO2026.047492
Countdown (A') + GRPOBackbone=Qwen2.5-7B, Generation budget=4096 tokens, Training Data=Self-revised answers (A'), Optimization=GRPO2026.047297
Countdown (A) + GRPOBackbone=Qwen2.5-7B, Generation budget=4096 tokens, Training Data=LLaMA-70B teacher answers (A), Optimization=GRPO2026.047197
Countdown (A')Backbone=Qwen2.5-7B, Generation budget=4096 tokens, Training Data=Self-revised answers (A')2026.047197
Countdown (A)Backbone=Qwen2.5-7B, Generation budget=4096 tokens, Training Data=LLaMA-70B teacher answers (A)2026.046996